On 8 April, 2018 at 18:47:33 UT a full moon bright bolide exploded over Hungary. Lots of meteorological camera caught the light of the fireball. Fortunately, three dedicated meteor camera also could observe the atmospheric trajectory. One of them was directly under the final phase of the fall and was able to take great pictures about it. The preliminary calculation shows that this event produced meteorite fall in Croatia.


Initial data

As always, I tried to collect every online reachable picture about this event. Among the first there was a great photo of Landy-Gyebnár Mónika which has been published on Facebook (Figure 1)

Figure 1 – The bolide’s photo from Veszprém, Hungary (Landy-Gyebnár Mónika’s picture)

On the same day another lucky catch turned up, Pócsai Sándor’s picture about the very end of the fall from Dávod. (Figure 2) The photo’s fine resolution was a great help to measure the end of the trajectory accurately.

Figure 2 – Pócsai Sándor’s photo about the fireball form Dávod, Hungary.

After a thorough search I could find numerous snapshots about the fireball (or at least its trail) among meteorological camera pictures. Because of licensing issues I can’t publish any of them here, but this does not prevent the scientific use and measurement of the images.

Three dedicated meteor cameras also could observe the event (Figure 3) moreover one of them was directly under the bolide’s flight path. This camera’s video served a good opportunity to detailed measure the formed debris cloud’s size and deceleration.


Figure 3 – The bolide’s snapshot from Sárrét, (Slovakia); Soroksár, (Hungary) and Becsehely, (Hungary)

The picture of Sárrét contains only the beginning part of the fall. Soroksár’s picture didn’t include the brightest phase of the fall, so the end of the trajectory is missing because of software issue of Metrec. Becsehely’s camera (also running with Metrec) somehow missed the first half of the fall, but could catch the brightest phase – for that reason highly saturated – and the fragmentation.



I have seven observations all around the meteor trajectory, four of them are calibrated manually and three was made by dedicated meteor cameras.

Working with Metrec’s data I noticed that the imprinted timestamps of the video frames were shifted, I had to take this into account when I was counting with them.

I also had to manually measure begin- and endpoints in UFOAnalyzer, (SonotaCo, 2009) because the software calculation depends on detection’s thresholds omitting frames, especially from the beginning of a fall. I used UFOOrbit’s (SonotaCo, 2009) import function to deal with the measured points.

Figure 4 – UFOOrbit calculated trajectory based on seven calibrated observations.

The meteor started its luminous path at 88.7 km (+/- 1.5) reaching the atmosphere with 29.5 degree (+/- 0.2) inclination. It flew with an average speed of 15.7 km/s (+/-0.5) from Kapolcs (Hungary) to Cvetkovec (Croatia) during more than 6 seconds. Above Becsehely reaching its peak brightness around -12 magnitude (+/-0.8) the fireball’s body fell apart and formed a 6 km long cloud of stone. The meteor’s pieces greatly decelerated at this phase, below 4.8 km/s. I could count the deceleration from frame by frame measurement of the Becsehely’s video in different heights of the fall. It also contains information about the initial velocity which was greater than 20.7 km/s. Luckily, the end of the trajectory was caught on fine resolution photos, so its accuracy is greater than the beginning. The last bits of the body could penetrate 27.2 km (+/-0.7) deep into the Earth’s atmosphere.

Table 1 – Measured velocity at different heights from Becsehely.

Height (km) Velocity (km/s)
63 20.71
31.6 5.08
29.9 4.846
29.6 4.8



I used the three dedicated meteor cameras’ observations to calculate the orbit of the fireball with the help of UFOOrbit, taking into consideration the deceleration.  I matched the measured values with already known fireballs’ speed curves and manually changed the meteor velocity for the deduced entry value (21.5 km/s) in the imported data. I would draw attention to the fact that without error spread calculations the resulted orbit is just a rough estimate. Strangely, the resulted orbit – within its error boundaries – is very near or intersects the orbit of Mars.

The resulting orbital elements are:

  • α = 246.6°
  • δ = +51.9°
  • a = 1.3 A.U.
  • q = 0.945 A.U.
  • e = 0.265
  • ω = 223°
  • Ω = 18.6°
  • i = 31.9°

Figure 5 & 6 – UFOOrbit calculated orbits with subtle differences between the observations.


Light and mass

After the event I found several visual observations online, in general they compared the fireball’s brightness to the full Moon. Especially those who were under the final phase of the fall.

As seen above, we have some very good photos about it, but because of the unique settings and physical configuration of the machines, it is difficult to determine a reliable brightness from them.

Metrec isn’t the finest tool either to measure precise light curve of a meteor. In this case, before the brightest phase – at around -3 magnitude – the software couldn’t follow the meteor’s trajectory and calculate its brightness because the highly saturated images. I had to estimate its peak brightness with the aid of an old picture about the full moon with the same camera. It was definitely in the same category or higher from Becsehely.

I calculated the photometric mass from the basic parameters of the event, absolute magnitude, velocity and zenith angle. (Jones et al., 1989) Insufficient knowledge about the brightness and the ablation coefficient increased the error margins greatly. The original mass was 1500kg (+/-1000), which corresponds to a one meter sized spherical body, assuming density of ordinary chondrite. After reaching the trajectory’s terminal point, a total mass of about 1 kg began its dark flight.


Figure 7 – Solid pieces in the wake at 29 km high.


Dark flight and strewn field

According to observations – while watching TV – people came out to the sound of explosion on the western part of the country. There was a double sonic boom which sounds like a distant thunder. Knowing this, seen the calculated residual mass and the deep penetration into the atmosphere there is a decent chance of some meteorites reached the ground.

I used a self-developed program called MetLab to calculate dark flight and the formed strewn field. Wind and atmospheric data can be retrieved from the University of Wyoming (Department of Atmospheric Science) website. In this case Zagreb’s radiosonde measurement was 70 km away from the terminal phase of the fall. I started the Monte Carlo simulation with 100 pieces of 100-300 grams meteorites from the last three km of the trajectory assuming density of ordinary chondrite. (Brown circles) After that I added another 100 pieces with any known errors. (Red circles)

Figure 8 – The calculated strewn field is in a forested area among rural villages in Croatia.



SonotaCo (2009). “A meteor shower catalog based on video observations in 2007-2008”. WGN, Journal of the International Meteor Organization, 37, 55–62.

Jones J., McIntosh B. A. and Hawkes R. L. (1989). “The age of the Orionid meteoroid stream”. Monthly Notices of the Royal Astronomical Society, 238, 179–191.

Department of Atmospheric Science, University of Wyoming, http://weather.uwyo.edu/upperair/sounding.html